Technologic and Nontechnologic Barriers to Implementing Behavioral Health Homes in Community Mental Health Settings During the COVID-19 Pandemic Open Access

People living with serious mental illnesses (SMIs) during the COVID-19 pandemic have experienced elevated mortality rates when compared to the general population, both from COVID-19 infection and all-cause mortality.^1–4  Prior to the COVID-19 pandemic, mortality rates for persons with SMIs have been documented to be 2 to 3 times higher than for the overall population,^5,6  mostly due to cardiovascular disease (CVD).⁷  The reasons for this excess morbidity and mortality are numerous and likely related to social determinants of health, biological factors, stigma, mental and physical health system challenges, and broader health and social policy issues.⁸ 

Behavioral health homes have been used in public sector mental health programs as a way to deploy collaborative care to improve physical health for those with SMIs. Established under the Affordable Care Act, additional funding was provided to community mental health programs to deliver comprehensive services akin to a “medical home model” for persons with SMIs, including physical health care in mental health settings to persons with SMIs.⁹  In particular, collaborative care, an evidence-based practice that could enhance the impact of behavioral health homes, has been shown to improve physical and mental health outcomes for persons with SMIs at little to no net health care costs.¹⁰  Collaborative care as operationalized in behavioral health homes typically seeks to integrate physical health care in a manner that is patient centered and coordinated, and to do so from within the community mental health setting.¹¹  This model assumes that discontinuities in care may happen owing to lack of integration in services and information flow within and between mental and behavioral health systems. Behavioral health homes operate within community mental health organizations, act as a liaison between mental health and physical health spheres, and may deliver some physical health care directly to mental health client organized around a patient registry (eg, monitoring blood pressures and blood glucose levels). Given the multiple symptom burden and functional limitations experienced by many with SMIs, health home services may include accompaniment of the individual to medical and other health care visits by a frontline provider such as a nurse, social worker, or care manager.

However, during the COVID-19 pandemic, mental health programs faced new challenges in delivering care to this vulnerable group, notably with the shift to acute care for the pandemic and subsequent delays in behavioral health and physical health (eg, preventative and chronic illness care) and the rapid pivot to virtual care and other remote-based technologies.¹²  Many services not directly related to the pandemic, including ongoing mental and physical health care, were not available, and for many individuals with SMIs, resulted in gaps in needed mental and physical health care.¹³ 

There has been little documentation of the specific barriers and lessons learned among behavioral health homes to implementing quality care for individuals with SMIs during the COVID-19 pandemic. Understanding how the pandemic impacted these services and the strategies frontline providers used to address these challenges can inform better implementation and sustainment of collaborative care for persons with SMIs and other chronic conditions.

The purpose of this study is to describe barriers to implementing behavioral health homes, experienced by community mental health workers in light of the COVID-19 pandemic, and the strategies used to address these challenges. In particular, this study sought to elucidate barriers and opportunities identified by frontline providers of services for persons with SMIs to ultimately inform improvements to the behavioral health home–based collaborative care model.

This qualitative study was conducted in community outpatient mental health programs participating in the Michigan-Maryland center that is part of the National Heart, Lung, and Blood (NHLBI)–funded Disparities Elimination through Coordinated Interventions to Prevent and Control Heart and Lung Disease Risk (DECIPHeR) Alliance. The goal of the DECIPHeR network of studies is to address known health inequities in CVD care and outcomes for a variety of populations.¹⁴  This study uses interview data collected in the developmental phase of the project to understand local implementation contexts, including anticipated barriers and facilitators, to tailor the intervention and implementation strategies. While not an explicit part of the interview guide, COVID-related topics surfaced as prominent themes.

Community mental health programs with behavioral health homes in Michigan and Maryland were invited to participate in the larger DECIPHeR initiative. Site directors were contacted and asked to provide names of providers at their facility to interview. Providers eligible to be interviewed included physicians, nurses, social workers, and frontline community services workers (eg, psychiatric rehabilitation workers, service coordinators) who were currently employed at the site.

This qualitative study reports methods and results in accordance with the guidelines presented in the Consolidated Criteria for Reporting Qualitative Research.¹⁵  To recruit participants, our study team set up introductory meetings with site leaders to provide an overview of the project and describe the anticipated time commitment. The study team then recruited clinicians and staff directly through information sheets, presentations at staff meetings, and when permitted by site leadership, direct emails to staff.

The study was reviewed and approved by the institutional review board (IRB) at the Johns Hopkins University School of Medicine, which served as the study’s single IRB for both Michigan and Maryland sites. A waiver of documentation of consent was obtained. Participants were informed of the overall purpose of the research, which was to gather information relevant to implementing health home services in their community mental health organizations. All interviewees provided consent to be interviewed and had their understanding of the study verbally confirmed. There was no prior established relationship to participants.

Interviews were conducted between January and June 2022 by a team of 7 trained interviewers who were part of the study team. Interviewers held bachelor’s, master’s, and doctoral degrees. The interview guide was developed by the study team, based on the Consolidated Framework for Implementation Research (CFIR).¹⁶  The CFIR is a theory-based framework used to assess determinants across multiple domains and levels understood to be important in implementation of innovations in health care. The CFIR was chosen as our initial framework for this study given its wide use and comprehensive description of the complex interactions of providers, systems, and environment with the introduction of new practices, especially within health care organizations and among populations with complex co-occurring conditions.¹⁷  The questions centered around how an evidence-based collaborative care model could be implemented in sites’ current behavioral health homes and potential barriers and facilitators. Questions about key domains thought to be important to implementation processes in the health home context were identified and iteratively modified following the first several interviews. Interviewers met throughout data collection to discuss any insights or modifications to the interview guide, and to assure consistency in interviewer approach. All interviews were conducted online via Zoom. Audio recordings were made during interviews and transcribed by using a professional transcription service. There were no others present during interviews.

Analyses were conducted by EMW, a White female who is PhD prepared in mixed methods/qualitative research and implementation science, and BO, a biracial female who is a doctoral student in health infrastructures and learning systems and a content expert in community mental health organizations and practice implementation. We first indexed all interviews by using the CFIR-informed code structure as part of the larger study aim to assess barriers and facilitators to implementing IDEAL Goals. An inductive approach was used to further investigate the study questions.¹⁸  We created initial categories from the data related to the impact of COVID on behavioral health home services. Categories were refined and combined in a second and third round of coding. An analytic cross-case matrix of categories was created, and illustrative quotes were added.¹⁹  Final themes regarding COVID-related barriers, facilitators, and strategies were created and displayed in tabular format. Owing to the large number of barriers, and very limited amount of data discussing facilitators, we then focused our analysis on COVID-related barriers to health home implementation, as well as anticipated barriers to implementing IDEAL Goals, and strategies used to overcome them.

An initial read of interview transcripts suggested that pandemic-related changes in implementation climate were highly salient. We then used MAXQDA’s (2022.2.0) autocode feature to index any text segments with the keywords pandemic, covid, covid-19, and coronavirus. Following a review of the first several cases, we added the keywords virtual, zoom, video, and telehealth. Next, we coded and conducted a thematic analysis in keeping with the method recommended by Braun and Clarke,²⁰  which involves an initial coding of the data, preliminary search for themes and second round of coding, review of initial themes, defining and naming finalized themes, and contextualizing themes initially coded, with BO double-coding the data, using the taxonomy presented in Table 1. Throughout the process we recorded illustrative quotes, and EMW and BO discussed code and theme refinement. A health home nurse and a health home administrator read and provided feedback on the manuscript as a member checking activity to increase the internal validity of the work.²¹  Stata/SE 17.0 was used to prepare the quantitative demographic variables.

Overall, 94 interviews among 112 community mental health staff were conducted across 21 mental health program sites, for a total participation rate of 84%. Of the interviews conducted, 72 containing information about COVID-related challenges to operating their health homes were included in this analysis. Demographic and workforce characteristics of the sample are presented in Table 2; 94% of the staff respondents were female and 19%, Black. Most participants had a bachelor’s degree and held a frontline mental health staff position. Thirty-six (n=50%) had some level of administrative responsibility, and 26% (n=19) were nurses.

Several barriers were identified in the operations of behavioral health homes that were attributed to pandemic-era changes at the individual mental health client, physical health system, and mental health system levels. Themes related to these barriers appeared to be divided into “technologic” and “nontechnologic” barriers, and thus are conceptualized and reported as such (Table 1). Respondents also discussed specific strategies that were used to attempt to overcome technologic barriers presented by COVID. Owing to space limitations, we have reported quotes representative of selected themes in Table 3.

Mental health staff reported that clients struggled both with access to technology and with being able to use it effectively to engage in virtual health home services (Table 3, row 1). Lack of cell phones, computers, and internet service were commonly noted, as well as trouble understanding how to use new technologies to access physical health care. Several strategies were used to tailor interventions to better work with client needs around these technologic access and skill problems. Grant and agency funding was used to purchase and distribute cell phones, tablets, and, in 1 case, cellular data plans. Respondents at many sites also attempted to coach clients on use of technology, with varying success (Table 3, row 2). Some frontline mental health staff went to clients’ homes to operate their tablets for them so that they could attend virtual medical appointments.

Clients also often had difficulties interpersonally engaging with mental health staff via modes of service delivery that were not in-person and relied heavily on technology (Table 3, row 3). For example, one health home nurse described how she often needed to have several calls with a client in order to complete a brief virtual health home service but was able to connect in-person for much longer (Table 3, row 4). Staff reported using a variety of strategies, with varying levels of success, to address this hesitancy, from providing reassurance to suggesting clients keep their cameras off while participating in virtual exercise classes (Table 3, row 5).

Many respondents reported trouble translating physical health interventions to an online format. This was particularly true for exercise classes, healthy cooking classes, and interventions relying on paper handouts (Table 3, row 6). Some staff reflected on the difficulty they foresee in preparing to implement a new CVD risk reduction intervention, given the current virtual care-heavy environment at the time of the interview (Table 3, row 7).

Health home staff also struggled with reduced ability to monitor CVD risk factors such as blood pressure (Table 3, row 8). Some community mental health programs attempted to address these problems related to obtaining up-to-date vital signs, weights, glucose readings, and other metrics for clients by distributing self-monitoring equipment or doing more home visits (Table 3, row 9).

A final mental health system barrier related to the new telework environment at many community mental health agencies regarding the perception that online and virtual trainings for both health home and non–health home topics are not as effective as in-person trainings (Table 3, row 10).

Many mental health staff perceived that their clients have struggled with a decrease in positive health behaviors due to pandemic-era isolation, and an increase in complexity of medical needs, possibly due to COVID infections and COVID isolation–facilitated social anxiety. Several staff members discussed issues around weight and eating during isolation (Table 3, row 11). Some clients also appeared to be struggling with new-onset physical health symptoms that were substantially greater than before the pandemic began (Table 3, row 12). In addition to these new physical health struggles, staff members perceived that clients often became more reluctant to return to the clinic, even when the opportunity arose, owing to having experienced an increase in social anxiety (Table 3, row 13).

A few COVID-related barriers related to challenges in the physical health care system. Mental health staff experienced a reduced level of information sharing and collaboration from primary care physicians and specialists, something they reported was already one of their biggest challenges (Table 3, row 14).

Within their mental health organizations, several respondents noted that some efforts to improve organizational processes around tracking of physical health measures and implementation of services to address CVD risk in their clients had been discontinued and not restarted. These included smoking cessation and smoke-free office efforts, an initiative to provide healthier food options to participants of day programs, and tracking of weights, blood pressures, and other measures (Table 3, row 15). One respondent also described an abandoned effort to develop closer relationships with primary care providers (Table 3, row 16). While some efforts to help consumers engage in physical exercise were moved to a telehealth format, others were not able to (Table 3, row 17).

Several additional barriers relating to the ability of health homes to address CVD risk factors were challenges experienced within both physical and mental health systems. Inability to attend medical appointments to advocate for clients owing to mental and/or physical health system policies, canceled programs and services, and lack of adequate staffing levels in both systems posed substantial barriers to the ability of health home and frontline mental health staff to helping their clients address CVD risk factors. At some agencies, smoking cessation programs were cancelled because public health departments that provided nicotine replacement therapy medications discontinued their programs (Table 3, row 18). Other respondents described how the inability during the pandemic to attend medical appointments with clients decreased how well they were able to bridge the gap between the community mental health and physical health care systems by building relationships with primary care providers (Table 3, row 19).

The perspectives of frontline mental health professionals, health home nurses, and community mental health administrators operating their behavioral health homes during the COVID-19 pandemic provide valuable insights into how evidence-based practices to improve physical health may be implemented under less than ideal conditions within mental health care systems. The experiences of those interviewed for this study represent the unique perspectives of health care professionals attempting to provide care for a medically vulnerable group of people during what is likely the most profound and abrupt change in the health care system during most health care professionals’ lifetimes.²²  Specifically, several issues related to health information technology (IT), as well as how to systematically address the long-term physical health consequences of living with SMIs during the COVID era, may warrant special attention from within an integrated collaborative care framework embedded within behavioral health homes.

One long-observed vulnerability within community mental health systems is the lagging behind in adoption of health IT for both providers and clients, compared to physical health care systems.^23–25  Health IT issues frequently rise to the forefront of analyses of barriers in community mental health care improvement efforts.^26–28  In a recent Commonwealth Fund Issue Brief, Docherty and colleagues²⁹  described how increasingly sophisticated implementation of integrated care components within mental health systems is either constrained by, or greatly facilitated by their health IT, and that the technologic challenges brought on by COVID magnified this. Our data support that integrated care-related initiatives around issues such as health measurement and appointment tracking can be fragile without adequate IT infrastructure.

This study also highlights that the relatively poor physical health status of people living with SMIs^30–32  was often made worse during COVID.³³  Existing barriers to care for persons with SMIs may have been further exacerbated by the COVID lockdowns, which restricted mental health providers from accompanying their clients to in-person physical health care visits, a part of care management for persons with SMIs who owing to their illness may have trouble navigating the fragmented health care system to begin with. While the contributors are likely multifactorial, the issues are likely to be responsive to careful and rigorous implementation of what we know already works: well-coordinated, evidence-based collaborative care that takes clients goals and preferences into account. Mental health systems planners and their general medical system counterparts might consider enhancements to current collaborative care interventions that explicitly and systematically assess clients for long-term negative consequences of COVID-19 infections, delayed medical care, disrupted linkages to primary care providers, and possibly risk of disconnection from the physical health care system due to continued difficulties some clients may experience with new reliance on physical health system telehealth. Clients facing poverty will be particularly vulnerable to being unable to access these care systems, and learning from those mental health sites that were able to effectively remove barriers to securing technologic devices for those in need may be particularly fruitful. Future research that uses a sociotechnical systems lens to seek a rich understanding of how social, technologic, and environmental factors intertwine to use technology to either facilitate or hamper quality of care in the community mental health setting should be prospectively explored.³⁴ 

Despite the large interview sample and in-depth interviewing, this study has limitations that warrant consideration. First, while several different types of community mental health frontline and administrative staff were interviewed for this study, the perspectives of clients of health home services are not represented and should be addressed by future work. Further, while the participation rate in this study was high (84%), reasons for refusal are not available. Third, since discussions about the impact of COVID on health homes occurred frequently yet spontaneously, this qualitative study precludes estimation of how frequently each barrier is present within community mental health programs. It is also possible that an availability heuristic colored staff members’ understanding of the importance of COVID-era systems changes in creating barriers. Finally, our study likely included workers who were viewed as being good candidates to speak on organizational functioning and, thus, had longer tenures than average.

In the wake of the COVID-19 pandemic, frontline mental health providers caring for persons with SMIs in community mental health programs experienced multilevel barriers to delivering behavioral health home services including collaborative care. They sought to overcome these barriers through technology, notably by helping clients use virtual care. At the same time, system-level changes wrought by the pandemic, including potential delays in care as well as labor shortages due to the shift to acute care settings, may have eroded progress these behavioral health homes were making to better integrate physical and mental health care for persons with SMIs. Technologic innovations brought on by COVID have increased access to care for some, but risk exacerbating health inequities for others owing to the potential of the “digital divide” experienced by persons with SMIs.³⁵  Additional strategies that address other structural barriers in addition to technologic barriers may be warranted to enable behavioral health homes to provide more effective collaborative physical and mental health care to persons with SMIs.

We would like to acknowledge funding for this work from the NHLBI UG3 HL154280. The views expressed are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs.